INDEX

Explanations

phrases where someone is reporting or communicating information

oai_token-act-pair · gpt-3.5-turbo

the word "the."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FIELD

-0.66

hement

-0.66

arettes

-0.62

sterdam

-0.62

ngth

-0.62

ItemThumbnailImage

-0.60

ãĥĥ

-0.60

untled

-0.60

edIn

-0.60

 Magicka

-0.60

POSITIVE LOGITS

 tale

0.87

 interviewer

0.75

 same

0.74

 latter

0.70

 story

0.70

 Crimson

0.69

 simplest

0.69

 latest

0.69

 voic

0.67

 commencement

0.65

Activations Density 0.178%

phrases where someone is reporting or communicating information

the word "the."

No Comments

No Known Activations